From 27c8079fded40c4d4345acd5aee489d29bb14470 Mon Sep 17 00:00:00 2001
From: Ingo Schwarze <schwarze@openbsd.org>
Date: Thu, 10 Apr 2014 02:46:21 +0000
Subject: Next speed optimization step for the new apropos(1). Split manual
 names out of the common "keys" table into their own "names" table.  This
 reduces standard apropos(1) search times (i.e. searching for names and
 descriptions only) by typically about 70% for the full /usr/share/man
 database. (Yes, that multiplies with the previous optimization step, so both
 together have reduced search times by a factor of more than six.  I'm not
 done yet, expect more to come.)

Even with the minimal databases built with makewhatis(8) -Q,
this step still reduces search times by 15-20%.  For both cases,
database sizes and build times hardly change (+/-2%).
---
 mansearch.c | 41 +++++++++++++++++++++++++++--------------
 1 file changed, 27 insertions(+), 14 deletions(-)

(limited to 'mansearch.c')

diff --git a/mansearch.c b/mansearch.c
index 3d070026..b2c7fe6d 100644
--- a/mansearch.c
+++ b/mansearch.c
@@ -1,4 +1,4 @@
-/*	$Id: mansearch.c,v 1.26 2014/04/09 21:50:08 schwarze Exp $ */
+/*	$Id: mansearch.c,v 1.27 2014/04/10 02:46:21 schwarze Exp $ */
 /*
  * Copyright (c) 2012 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -221,7 +221,7 @@ mansearch(const struct mansearch *search,
 				SQL_BIND_BLOB(db, s, j, ep->regexp);
 			} else
 				SQL_BIND_TEXT(db, s, j, ep->substr);
-			if (0 == (TYPE_Nd & ep->bits))
+			if (0 == ((TYPE_Nd | TYPE_Nm) & ep->bits))
 				SQL_BIND_INT64(db, s, j, ep->bits);
 		}
 
@@ -504,6 +504,12 @@ sql_statement(const struct expr *e)
 		    ? (NULL == e->substr
 			? "desc REGEXP ?"
 			: "desc MATCH ?")
+		    : TYPE_Nm == e->bits
+		    ? (NULL == e->substr
+			? "id IN (SELECT pageid FROM names "
+			  "WHERE name REGEXP ?)"
+			: "id IN (SELECT pageid FROM names "
+			  "WHERE name MATCH ?)")
 		    : (NULL == e->substr
 			? "id IN (SELECT pageid FROM keys "
 			  "WHERE key REGEXP ? AND bits & ?)"
@@ -525,8 +531,9 @@ sql_statement(const struct expr *e)
 static struct expr *
 exprcomp(const struct mansearch *search, int argc, char *argv[])
 {
+	uint64_t	 mask;
 	int		 i, toopen, logic, igncase, toclose;
-	struct expr	*first, *next, *cur;
+	struct expr	*first, *prev, *cur, *next;
 
 	first = cur = NULL;
 	logic = igncase = toclose = 0;
@@ -569,6 +576,7 @@ exprcomp(const struct mansearch *search, int argc, char *argv[])
 			first = next;
 		else
 			cur->next = next;
+		prev = cur = next;
 
 		/*
 		 * Searching for descriptions must be split out
@@ -576,18 +584,23 @@ exprcomp(const struct mansearch *search, int argc, char *argv[])
 		 * not in the keys table.
 		 */
 
-		if (TYPE_Nd & next->bits && ~TYPE_Nd & next->bits) {
-			cur = mandoc_calloc(1, sizeof(struct expr));
-			memcpy(cur, next, sizeof(struct expr));
-			next->open = 1;
-			next->bits = TYPE_Nd;
-			next->next = cur;
-			cur->bits &= ~TYPE_Nd;
+		for (mask = TYPE_Nm; mask <= TYPE_Nd; mask <<= 1) {
+			if (mask & cur->bits && ~mask & cur->bits) {
+				next = mandoc_calloc(1,
+				    sizeof(struct expr));
+				memcpy(next, cur, sizeof(struct expr));
+				prev->open = 1;
+				cur->bits = mask;
+				cur->next = next;
+				cur = next;
+				cur->bits &= ~mask;
+			}
+		}
+		prev->and = (1 == logic);
+		prev->open += toopen;
+		if (cur != prev)
 			cur->close = 1;
-		} else
-			cur = next;
-		next->and = (1 == logic);
-		next->open += toopen;
+
 		toopen = logic = igncase = 0;
 	}
 	if (toopen || logic || igncase || toclose)
-- 
cgit v1.2.3